The D. melanogaster stocks were maintained and amplified using a standard diet (Fly Food B, LabExpress, Ann Arbor, MI). The diet was supplemented with yeast (Carolina Biological Supply, Burlington, NC). Third-instar larvae from the D. melanogaster w¹¹¹⁸ wild-type line were used in the experiments. The flies were kept in an incubator at 25°C and a 12-h light: dark photoperiod.

Symbiotic and axenic H. bacteriophora TT01 nematodes were used in the experiments. The nematode stocks have been maintained and amplified in the Eleftherianos lab for several years. For nematode amplification, final stage larvae of the greater wax moth, Galleria mellonella, were infected with H. bacteriophora infective juveniles on a 6 cm Petri dish. The insect larvae were maintained for a week in an incubator set at 25°C and a 12-h light: dark photoperiod. Then, the infected caterpillars were transferred to water traps, as described before (Heryanto et al., 2022). Fourteen days later, the new generation of infective juveniles was collected in a 50 mL flask, maintained in an incubator at 28°C, and closely monitored under a stereoscope (Tritech Research, CA).

Nematode infection experiments of D. melanogaster w¹¹¹⁸ larvae were performed using 96-well microplates (ThermoFisher Scientific, USA). First, each well was loaded with 100 μL of 1.25% agarose gel (Fisher Scientific, USA). Then, single larvae were transferred with a fine paintbrush to individual wells of the 96-well microplate. Approximately 100 infective juveniles of either symbiotic or axenic H. bacteriophora suspended in 10 μL of sterile water were added to each larva (20 larvae were infected per survival experiment). Following infection, the microplates were sealed with Masterclear real-time PCR film (Eppendorf, Enfield, CT) and ventilated by creating small holes. Larvae that escaped from the well were excluded from this calculation. Control experiments involved treatment of w¹¹¹⁸ larvae with 10 μL of sterile water only (uninfected controls). Three independent experiments with new batches of D. melanogaster larvae and H. bacteriophora symbiotic or axenic nematodes were performed on different days.

Nematode-infected and uninfected D. melanogaster larvae were collected at 24- and 48-h time points. Only live larvae were collected in 1.5 mL Eppendorf tubes and stored in a freezer at −80°C for further processing. Each sample involved 100 larvae. DNA extraction was performed using the Qiagen DNA extraction kit protocol (Qiagen, Germantown, MD). DNA concentrations for each sample were measured with the Qubit 4 Fluorometer (Thermo Fisher Scientific, Waltham, MA). Each DNA sample had a minimum concentration of 10 ng/μL in a total volume of 100 μL. The purity of the samples ranged from 1.45 to 1.91, and the DNA concentrations varied between 20 ng/μL and 100 ng/μL.

Amplicon libraries were prepared using Zymo Research’s Quick-16S kit with phased primers (341F – 806R) targeting the V3-V4 regions of the 16S rRNA gene. Following clean up and normalization, samples were sequenced on a P1 600cyc NextSeq2000 Flowcell to generate 2 × 301 bp paired end (PE) reads. Quality control and adapter trimming was performed with bcl-convert1 (v4.2.4).

All the raw sequence files of this study were submitted to the European Nucleotide Archive (ENA) (O'Cathail et al., 2025) with the study accession number PRJEB85826 (available at http://www.ebi.ac.uk/ena/data/view/ PRJEB85826).

DADA2 was used to filter and trim sequences, infer amplicon sequence variants (ASVs) and remove sequencing errors and chimeric sequences. Taxonomy assignment was performed using SILVA release 138.2 (Quast et al., 2013). ASVs belonging to Mitochondria and Chloroplasts were removed before proceeding with the analyses. Phyloseq package (version 1.42.0) was used to calculate alpha and beta diversity (McMurdie and Holmes, 2013). Alpha diversity was estimated using the Observed Species, Chao1 (Chao, 1984), and abundance-based coverage estimator (ACE) indices (Chao and Lee, 1992). Beta diversity was analyzed using Bray–Curtis distances (Bray and Curtis, 1957) and visualized with nonmetric multidimensional scaling (nMDS); significance was assessed by permutational multivariate analysis of variance (PERMANOVA). Taxonomic groups with significant differences in abundance among different groups were identified by Linear Discriminant effect Size analysis (LEfSe) (Segata et al., 2011) using the microbiome Marker package (version 1.3.2) (Cao et al., 2022). Upsets plot showing how many ASVs and Species were unique and how many were shared between the groups was generated using UpSetR (version 1.4.0) (Nusrat et al., 2019) and ComplexUpset (version 1.3.3) (Krassowski et al., 2020; Lex et al., 2014). Rarefaction curves were generated using the MicrobiotaProcess package (version 1.6.6) (Xu et al., 2023). The aforementioned analyses were performed using R version 4.4.2.

Grouping the samples according to their infection status revealed significant statistical differences across the different treatments. This was confirmed by a PERMANOVA test, where an R² value was 0.12 and a p-value was 0.044 (Figure 1). This indicates that there was meaningful variation between the infected and uninfected samples. However, when the samples were grouped based on the specific treatment conditions, D. melanogaster larvae infected with either symbiotic or axenic H. bacteriophora or larvae treated with water only (uninfected controls), there was no statistically significant difference between these subgroups (PERMANOVA: R² = 0.18; p-value = 0.089). This suggests that the infection status is a primary differentiating factor, which means that it is the key variable that determined statistically significant changes in the microbial population in the fly larval microbiome. However, the type of infection - whether by symbiotic or axenic nematodes - or the absence of infection in control samples did not significantly impact the variation in microbial diversity. These findings were presented in the NMDS plot, where the samples are grouped into two distinct clusters- one for the uninfected control group and one for the infected group. The clusters reflect that the infection status had a stronger influence on the microbial composition of D. melanogaster larvae than the specific type of infection or the uninfected control condition.

The Upset plot, which determines the unique and shared bacterial species among the three experimental groups (D. melanogaster larvae infected with either axenic or symbiotic H. bacteriophora and uninfected control larvae), shows that from the total of 389 bacterial species 115 were shared among the three treatments (Figure 2); the D. melanogaster larvae infected with axenic H. bacteriophora contain 68 unique bacterial species. As observed in Figure 3, the plot highlights how the mutualistic P. luminescens bacteria of H. bacteriophora impact the microbial community in the D. melanogaster larvae during infection with the symbiotic nematodes (nematode-bacteria complexes). Particularly, these results indicate that D. melanogaster larvae infected with symbiotic nematodes support a microbial community which is distinct from that found in larvae infected with axenic nematodes or that observed in uninfected individuals. As shown in Figure 3, this indicates that the presence of P. luminescens mutualistic bacteria in H. bacteriophora nematodes plays a pivotal role in shaping the microbial community and host-microbial environment during D. melanogaster larval infection.

We have found that the uninfected control larvae contain a higher relative abundance of certain bacterial taxa such as Pseudomonadota and Bacillota compared to larvae infected with either axenic or symbiotic H. bacteriophora (Figure 3). This result highlights the difference in the microbial profiles between the three experimental groups. However, despite the differences, the bacterial communities observed in both uninfected controls and nematode-infected larvae mainly consisted of Pseudomonadota bacteria, namely from the Alphaproteobacteria and Gammaproteobacteria classes. The abundance of Pseudomonadota across all experimental groups suggests that these bacteria are characterizing the core microbiome of D. melanogaster larvae, regardless of their nematode infection status.

Linear Discriminant Analysis Effect Size (LEfSe) was performed using the three treatment groups (infection with H. bacteriophora symbiotic or axenic nematodes, or uninfected larvae) as a Group, while the two time points were clustered together as a Subgroup (Figure 4). The analysis showed a total of 28 bacterial species that were differentially abundant among the treatments. While the larvae infected with H. bacteriophora axenic nematodes showed 18 significantly enriched biomarkers, 10 biomarkers were more abundant in larvae infected with H. bacteriophora symbiotic nematodes. Stenotrophomonas maltophilia, one of the identified bacterial species, was highly abundant in larvae infected with H. bacteriophora axenic nematodes compared to the other treatments. From this analysis, it could be inferred that the presence or absence of the symbiotic bacteria P. luminescens influences the microbial composition of D. melanogaster larvae during nematode infection. Particularly, axenic H. bacteriophora nematodes lead to a greater shift in bacterial population diversity and are enriched in certain taxa such as Stenotrophomonas maltophilia. This suggests strong affirmation that the symbiotic bacteria play a key role in modulating the fly larval microbiome.

When plotted on a heatmap, the results of LefSe analysis show the distribution of the 28 bacterial species in the D. melanogaster larval microbiome in the uninfected control samples and the samples of the two groups infected with either symbiotic or axenic H. bacteriophora nematodes at 24 and 48 h (Figure 5). A noticeable increase in the bacterial abundance in the fly larval microbiome was observed in both nematode-infected groups (symbiotic or axenic H. bacteriophora-infected groups) in comparison to the uninfected control group, with the most predominant being Stenotrophomonas maltophilia, Reyranella, Ensifer adhaerens, Sphingomonas leidyi, Caulobacter henricii, Sphingobacterium multivorum, Achromobacter, Brucella, Mesorhizobiun, Sediminibacterium, Rhizobiaceae, Xanthobacteraceae, and Chitinophagaceae. Between the two nematode infected groups, a higher abundance of Nubsella, Brevudimonas diminuta, Vibrionimonas, Pseudoxanthomonas mexicana, Delftia tsuruhatensis, Brevundimonas naejangsanensis, and Bosea vestrisii was observed in the larvae infected with axenic H. bacteriophora; the larvae infected with symbiotic H. bacteriophora nematodes seemed to be closer to the control samples. Within the axenic H. bacteriophora infection treatment, almost all bacteria found 24 h following infection were also found in the similar abundance at 48 h. Stenotrophomonas maltophilia was the species with the highest relative abundance in larvae infected with axenic H. bacteriophora, exhibiting a 5-fold (log10) increase compared to the uninfected control group.

Boxplots show the alpha diversity indices, such as the observed number of Amplicon Sequence Variants (ASVs), Chao1, and Abundance-based Coverage Estimator (ACE), which estimate the richness and diversity of microbial communities in the D. melanogaster larvae upon infection with either axenic or symbiotic H. bacteriophora (Figure 6). In terms of the number of ASVs and the other alpha diversity metrics, the boxplots show similar values for both types of nematode-infected larvae. This indicates that the overall microbial diversity of the D. melanogaster microbiome was similar regardless of the type of nematode infection (symbiotic H. bacteriophora containing the symbiotic bacteria P. luminescens and axenic nematodes without the symbiotic bacteria). The higher values of the alpha diversity indices recorded in the EPN-infected groups compared to the uninfected control group indicate that the microbial communities in the infected (either by symbiotic or axenic H. bacteriophora) larvae were more diverse, i.e., that the larval infection results in higher alpha diversities.

Future research should investigate the unresolved questions regarding the shift of the microbial community on the host immune system response and overall survival. Notably, exploring how the altered microbial composition in D. melanogaster larvae following EPN infection impacts the host immune activity, metabolic function and behavioral responses could provide essential insights into host–parasite interactions. Also, the effect of EPN infection on the mucosal immune reaction of D. melanogaster or other insects that might, in turn, shape the microbiota as well as the insect host mucosal responses that might interfere with the endogenous microbiota will form future research objectives. The current research focused on two time points, 24- and 48-h post-infection. Future studies could expand this approach to include additional time points to better understand how microbial communities evolve and their consequent impact on the host. Moreover, multiple species of parasitic nematodes could be incorporated into future research to understand how different EPN-host–microbe interactions stimulate microbial community shifts. Assessing various environmental parameters, such as temperature, could reveal how these factors influence the interactions between nematode infection and host-microbiome composition, which would propose a more comprehensive understanding of the ecological and environmental impact on host-microbial community dynamics.

Furthermore, the application of advanced high-throughput sequencing technologies could facilitate the identification of unique microbial taxa responsible for altering immune responses. This could lead to the innovation of novel therapeutic strategies that would target microbial communities. Integrating approaches such as transcriptomics and metabolomics could also provide profound insights into the molecular mechanisms between the host, the EPNs, and the microbial population. This would pave the way for improved strategies in managing parasitic nematode infections in agriculture and medicine. In conclusion, by expanding our knowledge of these complex relationships, future research can benefit from practical solutions to combat parasitic infections and improve human health.